Colleen M. Ennett

Prediction of extubation failure for neonates with respiratory distress syndrome using the MIMIC-II clinical database

Extubation failure (EF) is an ongoing problem in the neonatal intensive care unit (NICU). Nearly 25% of neonates fail their first extubation attempt, requiring re-intubations that are associated with risk factors and financial costs. We identified 179 mechanically ventilated neonatal patients that were intubated within 24 hours of birth in the MIMIC-II intensive care database. We analyzed data from the patients 2 hours prior to their first extubation attempt, and developed a prediction algorithm to distinguish patients whose extubation attempt was successful from those that had EF. From an initial list of 57 candidate features, our machine learning approach narrowed down to six features useful for building an EF prediction model: monocyte cell count, rapid shallow breathing index, fraction of inspired oxygen (FiO2), heart rate, PaO2/FiO2 ratio where PaO2 is the partial pressure of oxygen in arterial blood, and work of breathing index. Algorithm performance had an area under the receiver operating characteristic curve (AUC) of 0.871 and sensitivity of 70.1% at 90% specificity.

Performance evaluation of various storage formats for Clinical Data Repositories

The performance of three different Entity-Attribute-Value (EAV) storage formats for Clinical Data Repositories (CDRs) is compared with regards to querying millions of data points from clinical sources to assess the amount of storage space the data use, the speed with which the data can be obtained, and the complexity of the queries required to retrieve the data. Performance results are presented that show that the hybrid EAV approach provides a nice balance of the simple and multi-data type formats.

Performance Evaluation of Various Storage Formats for Clinical Data Repositories

The performance of three different Entity-Attribute-Value (EAV) storage formats for Clinical Data Repositories (CDRs) is compared with regards to querying millions of data points from clinical sources to assess the amount of storage space the data use, the speed with which the data can be obtained, and the complexity of the queries required to retrieve the data. Performance results are presented that show that the hybrid EAV approach provides a nice balance of the simple and multi-data type formats.

Predicting respiratory instability in the ICU

Acute lung injury (ALI) and acute respiratory distress syndrome (ARDS) contribute to the morbidity and mortality of intensive care patients worldwide, and have large associated human and financial costs. We identified a reference data set of 624 mechanically-ventilated patients in the MIMIC-II intensive care database with and without low PaO2/FiO2 ratios (termed respiratory instability), and developed prediction algorithms for distinguishing these patients prior to the critical event. In the end, we had four rule sets using mean airway pressure, plateau pressure, total respiratory rate and oxygen saturation (SpO2), where the specificity/sensitivity rates were either 80%/60% or 90%/50%.

Usefulness analysis of a Clinical Data Repository design

The perceived usefulness of a Clinical Data Repository (CDR) prototype in a hospital setting was assessed by clinicians to determine whether they would find it helpful for their clinical and research work. The CDR automatically collects and stores clinical data in real time from patient monitoring devices, clinical information systems, laboratory systems, and the Health Records Department in a de-identified, easily extractable format for secondary uses. A secure online survey was distributed to physicians, research institute investigators, and research institute coordinators at the Childrens Hospital of Eastern Ontario (CHEO) through email. According to the survey responses, participants felt the CDR was a useful tool, showed interest in it, and thought it would be important to have for future work. To illustrate how the CDR could be used in a clinical setting we have provided a sample clinical application; a tool for engaging physicians and parents in discussion about the clinical progress and prognosis of infants in the Neonatal Intensive Care Unit (NICU).

Heuristics to determine ventilation times of ICU patients from the MIMIC-II database

Mechanical ventilation is an important life support tool for patients in intensive care units (ICU). For various research purposes related to patient hemodynamic and cardiopulmonary monitoring, it is important to know when a patient is on a ventilator. Unfortunately, the widely used MIMIC-II database contains results from user charted data, where the user did not always store ventilation on and off times explicitly and accurately. The resulting ventilation-related data are subject to error. Therefore, there are no simple rules to define ventilation times retrospectively for this dataset. Hence, we designed a simple set of rules to determine the ventilation times using multiple sources of mechanical ventilator-related settings and physiological measurements by expert heuristics. The rules worked well in comparison with nursing notes regarding ventilation events. We conclude that our rule sets for determining ventilation times may be useful in assisting with MIMIC-II database analysis.

Discrimination of Inconsistencies in Medical Data

Missing and erroneous values in patient cases can significantly impact the ability to perform biomedical research for identifying risk factors and causes of clinical events and disease progression. We present a framework that classifies the inconsistencies in a database to automatically process the data for research purposes. The goal is to improve the quality of medical databases by identifying the reason that data are missing, and automatically processing outliers and typographical errors to retain more information from a database. This framework presents an alternative to deleting cases with outliers and typographical errors, while also imputing relevant values into cases with missing data based on the reason that the data is missing, the source of the data, and the data type.